Research on English Writing Text Analysis and Personalized Teaching Material Recommendation System Based on BERT Model

With the continuous development of artificial intelligence technology, natural language processing (NLP)-based methods are more and more widely used in the field of education, especially in the teaching of English writing, and the demand for automated writing analytics and personalized recommendation systems continues to grow. However, most of the current writing analysis systems still rely on traditional rule-based methods or classical machine learning algorithms, which are still insufficient in semantic understanding and personalized support. Therefore, how to improve the intelligence of English writing analysis systems, especially in terms of grammatical error detection, semantic understanding and recommendation strategies, has become a key issue that needs to be addressed urgently. To this end, this study proposes a text analysis and personalized recommendation system for English writing based on the BERT model. The system utilizes the powerful semantic understanding capability of BERT to automatically detect grammatical, spelling, and semantic errors in students' writing, and combines it with a personalized recommendation algorithm to recommend appropriate learning resources based on students' writing performance and error types. The experimental results show that the BERT-based system outperforms traditional rule-based methods and classical machine learning models in several tasks, with the accuracy of grammatical error identification improved by 12.3% and the accuracy of semantic understanding reached 88.2%.